During the first 30 months of the original grant (R01 CA 092039-01A2), we have successfully contributed to our understanding of the genetic epidemiology of lung cancer and upper aerodigestive tract (UADT) cancer by identification of several genes that are very strongly associated with them, including CHEK2 and ADH1B. We have also demonstrated how genetic variants interact strongly with dietary and environmental exposures for these cancers. Given the enormous increase in genetic information over the last 3 years we plan to build on our initial results and comprehensively evaluate the role of genes in 5 specific pathways for these cancers. To test the robustness of the positive associations observed, we will conduct independent replication of findings in other large studies. We therefore propose a multistage study with the following specific aims: Stage 1 will involve a comprehensive evaluation of 5 candidate gene pathways among 2200 European case-control pairs of lung cancer and 1000 case-control pairs of UADT cancer, involving over 1500 informative variants as well as inclusion of biologically relevant variants. Stage 2 will involve rapid replication of important positive results in other large independent European studies including (i) EPIC lung cancer based on 1200 lung cancer cases and 2400 controls, and (ii) the 'ARCAGE'Western European study of 2000 case-control pairs of UADT cancer. The choice of variants passing from Stage 1 to Stage 2 will be based on hierarchical bayes approach incorporating genomic information such as sequence conservation. Important confirmed genes will be resequenced and further replicated in a third IARC study of UADT cancer from Latin American study comprising 2000 case-control pairs of head and neck cancer. As an important component of this proposal, we will conduct functional studies of the genes that are replicated including differential mRNA expression. All results will be made available to the research community of tobacco and alcohol related cancers via collaboration within 2 international consortia of lung cancer and head and neck cancers. Relevance of research to public health: Large genetic epidemiology studies have the potential to identify individuals at particularly high risk of developing cancer, as well as helping identify why these cancers develop. By incorporating multiple large studies of cancers related to tobacco and alcohol (specifically lung, and UADT cancer) we aim to provide knowledge that will inform future prevention efforts.